Please note that this is an author-produced PDF of an article accepted for publication following peer review. The definitive publisher-authenticated version is available on the publisher Web site c b a http://dx.doi.org/10.1016/j.ecolmodel.2007.10.047 April 2008, Volume212, Issues 3-4,Pages422-438 Ecological Modelling Gaines, al., 1998],[Lodge etal., 2002],[Kolar1998], [Allen andHumble, andLodge, 2002], [Sax and to detrimentalimpactson ecologyof wildstocks [Irz etal., 2004] and[Saltveit,2006]). However,thisstrategy isoften unacceptable as it leads FAO FisheriesTechnical 1988],[Pitcher andHart,1995], Papers294, 318pp.[Welcomme, 1988.InternationalTransfersofInlandFish Species.R.L., 1988Welcomme, (Welcomme, economic benefits, as biomanipulation stratagem andsustain recreational fisheries food tosustain increasing humanpopulation and poverty,enhancefish stocks, generate Introduction ofalienspecies inaquatic ecosystems isstimulatedbyincreasingdemand for Introduction between system components (i.e.fish, invertebra therefore, be consider as “immature”, sensu Od majority offishbiomass concentration is(55%). inTL3 Thefishery isconcentrated atTL3andcan, distribution of some of the fish functional groups from TL1 to higher TLs due tothe abundance of zo levels (TL) varies between 4.5 and 9.4%. Zooplankton food web, itisobserved that key aphytoplankton-based As considered. were compartments Fourteen ecosystem. this on framework, considered as aninitialstep insummarizing ecological and biological information, under a coherent quantitatively describe thepossibleimpact offishin An Ecopath model of the southern part ofLake Ki Abstract: 15601, emailaddress: *: Corresponding author : Maria Concepcion S.Vi d Keywords: are discussed. described and thepossible influence and role inthe ecosystem's functioning of introduced fishspecies © 2008ElsevierB.V.Allrightsreserved.
32607, 31326 Castanet Tolosan, France France Tolosan, Castanet 31326 32607, France Bessin Huppain, Republic of Congo ofCongo Republic UERHA, Département de Biologie, InstitutSupérieu Biologie, de Département UERHA, Department of Tropical Fisheries, Lab Fisheries, of Tropical Department Port-en- BP32,14520 deGaulle, duGénéral IFREMER, Avenue Halieutiques, des Ressources Laboratoire URBO, FUNDP, Rue de Bruxelles 61, B-5000 Namur, Belgium Belgium Namur, 61,B-5000 URBO,FUNDP, RuedeBruxelles Modeling trophicinteractionsinLake Maria ConcepcionS.Villanueva Food webstructure; Transfer efficiency; Invasion;Exoticfishes; LakeKivu [email protected] Moreau food sources arenot fullyutilizedas oratoire d’Agronomie, Environnement d’Agronomie, oratoire a, b, * b andJean-ClaudeMicha , MwapuIsumbisho vu, adeep African equatorial lake was constructed to
r Pédagogique de Bukavu, BP 854, Bukavu, People's People's Bukavu, 854, BP de Bukavu, Pédagogique r were observed asaresponseto competition.The tes, phytoplankton and detri um. Thedirectandindirecteffectsofpredation
llanueva, Tel.: +33 2315 15637; fax:+33 2315 Kivu:Whatroles thatfrailsecosystem structure([Wilcoveet plays a major role intransferring organic matter oplanktivores. Shiftsin food preferences and troductions inthisecosystem. Thisstudyis c , Boniface Kaningini d
et Ecotoxicologie, INP/ENSAT, BP INP/ENSAT, etEcotoxicologie,
transfer efficiencies per trophic Archive Institutionnelle del’Ifremer Archive Institutionnelle http://www.ifremer.fr/docelec/ do exotics play? tus) arequantitatively c , Jacques Archimer 1
25 2003; Gurevitch and Padilla, 2004; Didham et al., 2005; Arim et al., 2006), fisheries economy
26 (Mack et al., 2000; Pimentel et al., 2001) and recreation (Winfield and Durie, 2004).
27
28 Widespread introductions of non indigenous species have been categorized as a major
29 cause of natural species extinction compared to habitat fragmentation (MacDonald et al., 1989;
30 Lodge et al., 1998; Davis and Thompson, 2000; Allen and Humble, 2002; Sax and Gaines, 2003;
31 Gurevitch and Padilla, 2004; Didham et al., 2005; Arim et al., 2006) in both terrestrial
32 (Rejmánek and Richardson, 1996; McCann, 2000; Smith et al., 2000; Allen and Humble, 2002;
33 Guo et al., 2006; Lovett et al., 2006) and aquatic systems (Mills et al., 1993; Pitcher and Hart,
34 1995; Puth and Post, 2005; Latini and Petrere Jr., 2004; Dudgeon et al., 2006). Although in the
35 latter, biological invasions have been recognized as a persisting problem compared to pathologic
36 crisis in terrestrial ecosystems (Dudegeon et al., 2006).
37
38 Questions on success of exotics and the damaging impacts to native stocks at the
39 ecosystem level have fascinated many ecologists, such as Crawley (1987), Naeem et al. (2000)
40 and Kennedy et al. (2002). Ecological systems are extremely complex networks, consisting of
41 many biological species that interact in many different ways, such as mutualism, competition,
42 parasitism and feeding relationships. The latter can cause invasions, extirpations, and population
43 fluctuations of a species to dramatically affect other species within a variety of natural habitats
44 (Pimentel et al., 2001; Winfield and Durie, 2004). According to Hobbs (1989), successful
45 invasion in natural communities depend on species dispersal, establishment and survival with the
46 number of species per area established by immigration-extinction equilibrium. Success of exotics
47 also depends on tolerance and broad ecological demands, ability to adapt to habitat and
48 environmental conditions and r-selected life histories (Craig, 1992; Murichi et al., 1995).
2 49
50 Cases where introduction of exotics have been reported beneficial are rare in both
51 terrestrial (Schutzenhofer and Valone, 2006) and aquatic ecosystems (Gottlieb and
52 Schweighofer, 1996). Elevated biodiversity has been observed to increase resistance from
53 invasions in terrestrial and aquatic systems by creating insurance through functional redundancy
54 (Simberloff and Von Holle, 1999; Sax and Brown, 2000; Naeem et al., 2000; Kennedy et al.,
55 2002; Raffaelli et al., 2002; Stachowicz et al., 2002).
56
57 The importance of considering a trophic network approach is that it can elucidate feeding
58 relationships which occur between species in an ecological community and determine functional
59 roles of species groups in the ecosystem (Yodzis and Winemiller, 1999). Indeed, numerous
60 evidences suggest that food web structures are susceptible to a wide array of human activities,
61 including species introductions or invasions (Vander Zanden et al., 1999), habitat alteration
62 (Wootton et al., 1996), and global environmental warming (Petchey et al., 1999).
63
64 Quantitative trophic analyses at the ecosystem level were carried out in some African
65 Lakes where exotic species were introduced (Moreau, 1995; Moreau et al., 1993; 2001;
66 Villaneuva and Moreau, 2001). A similar approach has been carried out in other African lakes,
67 i.e. Lake Victoria (Moreau, 1995; Villanueva and Moreau 2001), Lake Naivasha (Mavuti et al.,
68 1996; Moreau et al., 2001) or Lake Kariba (Moreau, 1997), to determine the state of biologic
69 community alterations following fish introductions. As effects of fish introductions and its
70 exploitation on the community and ecosystem level are still unknown in Lake Kivu. The aim of
71 the present contribution is to study the food web structure, species interactions, role of exotics in
72 the ecosystem and compare these to observations in other tropical lakes where fish introductions
3 73 occurred. Understanding trophic links is crucial in predicting future impact of species invasion in
74 natural food web structure and functioning.
75
76 Material and Methods
77 Study site
78
79 Lake Kivu (Fig.1) has a surface area of 2 370 km² of which 1370 km² is a part of the
80 Congolese territory. An international aquatic system situated along the Congo-Rwanda border at
81 an altitude of 1 463 m. It is located between 1°30’ and 2°30’ latitude south and between 28°50’
82 and 29°23’ longitude east. It is a deep (maximum depth 490 m) equatorial lake with an average
83 water depth of about 240 m. The littoral area stretches not further than 50 m away from the
84 lake’s extensive (1200 km) shoreline (Van den Bossche and Bernascek, 1990; Verheyen et al.,
85 2003). It is a meromictic lake with deep relict hypolimnion where beneath lies a vast methane
86 gas reserves (Coulter et al., 1984; Snoeks, 1994). Permanent water stratification is observed:
87 anoxic below 60 m while the deeper part of the lake is methane saturated (Coulter et al., 1984;
88 Van den Bossche and Bernascek, 1990; Isumbisho et al., 2006; Sarmento et al., 2006). Annual
89 precipitation in the region is about 1 300 mm, relatively higher along the occidental than the
90 oriental side of the lake, which experiences virtually no variations in water level. The average
91 surface water temperature is about 24°C (Snoeks, 1994).
92
93 Several lakes of the East African Rift Valley are characterized by a deep pelagic zone
94 which is colonized by abundant native small pelagic fish (Coulter et al., 1984; Lowe-McConnell,
95 1993). A well documented exception among these lakes is Lake Kivu. Compared to other lakes
96 of the Rift Valley the fish diversity is relatively poor with only 26 endemic species belonging to
4 97 the Cichlidae, Clariidae, Cyprinidae and Clupeidae families (Hanek et al., 1991; Snoeks, 1994).
98 The Cichlids are the most represented with 17 endemic haplochromines (De Vos et al., 2001).
99
100 Exotic fishes were introduced to increase biodiversity and productivity of the lake
101 (Welcomme, 1988). Fish stocking in Lake Kivu dates back in the 1950s where two cichlids,
102 Oreochromis macrochir (Boulenger) and Tilapia rendalli (Boulenger), were introduced due to
103 the renowned ecological plasticity of these species (Chapman et al., 2003; De Vos et al., 2001).
104 Two endemic sardines of Lake Tanganyika, Limnothrisssa miodon (Boulenger) and Stolothrissa
105 tanganyicae (Regan), were then simultaneously introduced in 1959 (Van den Bossche and
106 Bernascek, 1990; Spliethoff et al., 1983) to occupy the pelagic zone (90%). S. tanganyicae,
107 however, was not able to adapt to the local conditions in the lake (Hauser et al., 1995).
108
109 The Lake Kivu fishery is predominantly artisanal (Van den Bossche and Bernacsek,
110 1990; Hanek et al., 1991; de Iongh et al., 1995) which is similar to other East African Lakes
111 (Pitcher and Hart, 1995; Preikshot et al., 1998). In terms of the fishing activity, fishery in the site
112 considered is the most important in the Congolese sector (Hanek et al., 1991). Annual production
113 is generally observed higher in the Rwandese sector where fishing activities are more active and
114 developed. At the zone considered in this study annual production in 1990 represented 20% of
115 overall production (Hanek et al., 1991; Marshall and Mubamba, 1993). Fishermen operate with
116 various fishing gears depending on season, investment level, fishing areas and species targeted.
117 A specific fishery, trimaran, uses light attraction and liftnet and selectively targets L. miodon and
118 planktivore haplochromines (Van den Bossche and Bernacsek, 1990; Hanek et al., 1991; de
119 Iongh et al., 1995; Kaningini et al., 1999). Beach seines capture mainly the benthopelagic
120 haplochromines, but have been observed to accidentally catch other species such as L. miodon.
5 121 Gillnets capture mainly tilapias although smaller mesh-sized (10 mm) nets are employed to trap
122 L. miodon and some haplochromines. Longlines target mainly Clarias species (Hanek et al.,
123 1991).
124
125 The lake is an international area shared by Rwanda (East) and RD Congo (ex-Zaire,
126 West). For this study, we considered the Bukavu basin of the Congolese sector (Fig. 1) which is
127 approximately 140 km², as this zone is better documented in terms of biological community
128 ecology and fisheries compared to the Rwandese sector. This zone also represents an important
129 socio-economic aspect (Hanek et al., 1991). It should be noted that parameters integrated in the
130 model were mainly estimated using data collected in this area.
131
132 Theoretical Approach
133
134 We used the Ecopath model (Christensen and Pauly, 1993; Christensen et al., 2005) to
135 construct a steady-state description of the Bukavu Bay. The model has already been used for
136 quantifications of food webs in different ecosystems to study the impact of fisheries for
137 management purposes (Pauly et al., 2003; Christensen and Walters, 2004b). It comprises a set of
138 simultaneous linear equations, one for each group under consideration, where the production of
139 the group is equal to the sum of all predation, non-predatory losses and export:
140 n P Q P 141 Bi ---- = ∑ Bj ---- DCji + Bi ---- (1-EEi) + EXi (1) Bi j = 1 Bj Bi 142
143 where Bi is the biomass of group i (in t km-2 fresh weight); P/Bi is the annual production/biomass
144 ratio of i equal to the total mortality coefficient (Z) in steady-state conditions (Allen, 1971); EEi
6 145 is the ecotrophic efficiency representing the part of the total production consumed by predators
146 or captured in the fishery or exported; Bj is the biomass of the predator group j; Q/Bj is the annual
147 food consumption per unit biomass of the predatory group j; DCji, is the proportion of the group i
148 in the diet of its predator group j; EXi, is the export or catch in fishery of group i, that is assumed
149 to be exploited in the fishery (Christensen et al., 2005).
150
151 In addition to balancing the model, Ecopath can be used to compute parameters and
152 indices corresponding to the food web characteristics. Some parameters that can be estimated
153 using the software are as follows:
154
155 a.) The group-specific omnivory index OI is computed as the variance of the TLs of each
156 predator’s prey groups (Christensen and Pauly, 1993) while the system omnivory index (SOI) is
157 computed as the average omnivory index of all consumers weighted by the logarithm of each
158 consumer's food intake Q (Christensen et al., 2005). It indicates the allocation of predator170
159 prey interactions linking each TL (Christensen and Walters, 2004a). Both OI and SOI indices
160 vary from 0 to 1, where a value close to 0 indicates high predatory specialization (feeding on one
161 trophic level only) and 1 indicates a maximum feeding versatility on several trophic levels.
162
163 b.) The connectance index (CI) is the ratio between the number of actual definite trophic
164 associations among all the groups and the theoretical possible number of connections, (N-1)² for
165 N groups, including consumption of detritus (Christensen and Walters, 2004a; Christensen et al.,
166 2005). This index is correlated with the maturity e.g. the level of evolution of the ecosystem, as
167 defined by Odum (1969), of the ecosystem because the food chain structure changes from linear
168 to web-like as a system matures (Odum, 1971).
7 169
170 c.) Niche overlap is measured by using a symmetrical index derived from the one proposed by
171 Pianka (1973) based on competition coefficients of the Lotka-Volterra equation (Volterra, 1931)
172 and derived from the Jaccard similarity index (Harris, 1968). It is used to describe mainly the
173 trophic aspect of niche partitioning. An index value close to 0 indicates that two groups have a
174 low resemblance in terms of food consumed and vise versa for a value close to 1 (Christensen et
175 al., 2005).
176
177 d.) Trophic aggregation per discrete TL, sensu Lindeman (1942), is based on an approach
178 suggested by Ulanowicz (1995). This routine facilitates calculation of flows per TL based on diet
179 compositions by reversing the routine for calculation of fractional trophic levels quoted above.
180 More particularly, the transfer efficiencies between the successive discrete trophic levels are
181 calculated as the ratio between the sum of the exports plus the flow that is transferred from one
182 trophic level to the next, and the throughput at this trophic level (Christensen et al., 2005).
183
184 e.) The gross efficiency of the fishery (GEF) is computed as the ratio between the total catch and
185 the total primary production in the system. The value will be higher for systems with a fishery
186 harvesting fish belonging mainly to low TLs than for systems whose fisheries concentrate on
187 high TLs. Therefore, this index may increase with fisheries ‘development’ as defined by Pauly et
188 al. (1998).
189
190 Model construction
191
192 The model was constructed using data collected from 2002 and 2003. These are
8 193 complemented by additional published works by Ulyel (1991), Snoeks (1994) and Kaningini et
194 al. (1999). For simplification purposes, species with similar ecological characteristics (i.e.
195 metabolism, diets and predators) were pooled together following the indications of Yodzis and
196 Winemiller (1999). In such a case, the biological characteristics of the most abundant species
197 were considered. A total of 14 groups were considered in this study and the ecological grouping
198 of biological assemblages is the following one.
199
- 200 Detritus: A standing stock of 165t.km ² (fresh weight, fw) was calculated using the
201 empirical equation of Pauly et al. (1993) based on an annual primary production estimated in the
202 system (see below) and a euphotic zone of 32 m. It is an input required to run the model
203 (Christensen and Pauly, 1993).
204
205 Phytoplankton and primary production: The dominant groups are diatoms (Nitzschia spp
206 and Fragilaria spp) and cryptophytes (Chroomonas spp and Rhodomonas spp). Chlorophyll-a
207 concentration showed clear seasonal variations and increases during the dry season, after deep
208 mixing in the basin of Bukavu (Ishumbisho et al., 2006; Sarmento et al., 2006). The average
-3 209 Chlorophyll-a content is 1.53 mg m for the area considered. It can be extrapolated over 60 m as
210 suggested by Isumbisho et al. (2006) and the resulting fresh biomass (32.1 t km-² fw) was used as
- 211 an input in the model. The annual primary production was measured at 273 g C.m ² by Sarmento
212 et al. (2006) and concords with annual range values (240 – 379 gC.m-²) indicated earlier by
213 Jannasch (1975), Van den Bossche and Bernacsek (1990) and Descy and Fourniret (1991). A
214 similar value was estimated using the Photosynthesis simulator of Capblanc and Dauta (1999).
-1 215 Assuming 1 g C is equal to 10 g fw (Sarvala et al., 1999). A P/B ratio of 85 yr was used as an
216 input in the model.
9 217
218 Zooplankton and secondary production: The zooplanktonic community is dominated by
219 copepods (Thermocyclops) and, to a lesser extent, cladocerans (Diaphanosoma) though seasonal
- 220 density variations can be observed (Isumbisho et al., 2006). The average B is 5.64 t.km ²
-1 221 estimated from Isumbisho et al. (2006) and an annual P/B value of 30 yr was used (Sarvala et
-1 222 al., 1999; Irvine and Waya, 1999). Q/B value of 180 yr was adopted from Sarvala et al. (1999).
223
224 Benthic fauna: Only limited information was available for this group. This includes
225 benthic deposit feeders such as the nematodes, ostracods, insects, bivalve mollusks, gastropods
226 and other benthic organisms which develop only along the littoral zone due to the relative
-1 227 steepness of the lake. A P/B value of 4.5 yr (Payne, 1986; Mavuti et al., 1996) was adopted and
-1 228 Q/B of 45.0 yr was assumed from a gross efficiency (GE or P/Q) value of 0.1 (Christensen and
229 Pauly, 1993). An input value of 0.900 for EE was used to estimate the biomass based on intense
230 predation noted from higher TL consumers.
231
232 Fish groups: When possible the B/P ratio was estimated from recently collected length
233 frequency distributions by using the FiSAT software (Gayanilo et al., 2002). In a first step, this
234 software was used to estimate the growth parameters of the von Bertalanffy growth function i.e.,
235 the asymptotic length (L∞) and the growth coefficient (K) which are needed for P/B computation
236 by reference to the length converted catch curve method. Otherwise, the predictive models of de
237 Merona (1983) or Fröese and Binohlan (2000) were employed to estimate these demographical
238 parameters. Natural mortality, M, was computed using the predictive formula of Pauly (1980).
239 The demographical data of the fish populations considered here are summarized on Table 1.
240
10 241 The food consumption per unit of biomass (Q/B) has been estimated in a few cases using
242 Maxims (Jarre et al., 1991), a software model based on the method of Pauly (1986) which allows
243 the computation of Q/B from an estimate of the daily food consumption of individual fish of a
244 particular size. Otherwise, Q/B was calculated using the multiple regression formula of
245 Palomares and Pauly (1998).
246
247 For most fish groups, local field data on diet composition were available. Additional
248 information was taken from Snoeks (1994) and Ulyel et al. (1991).
249
250 The biomass (B) of each fish group was estimated assuming equilibrium conditions, such
251 that:
252 B = Y/F (3)
- -1 253 where Y is yield in t.km ².yr and F is the coefficient of fishing mortality. F is the difference
254 between total and natural mortalities: F = Z – M, assuming that Z is equal to P/B as indicated by
255 Allen (1971).
256
257 Due to habitat preferences of the species under investigation, the littoral area which is
258 about 10% of the total surface of the lake (based on depth and bathymetry) was separated from
259 the pelagic zone. The biomass of each compartment was calculated according to habitat area. For
260 littoral groups, the biomass per km² as computed as an average for the whole lake was, then,
261 multiplied by 10 in order to express the concentration of this group in the littoral area.
262
263 Catch data and the proportion of each group in the total catch were obtained from various
264 sources (i.e., Van den Bossche and Bernacsek, 1990) including unpublished data (J-C. Micha,
11 265 FUNDP; B. Kaningini and M. Isumbisho, UERHA pers. comm.) recently collected as a part of
266 an on-going Belgium Project for development of a gillnet fisheries in the lake.
267
268 The following groups were considered
269
270 1°) Clarias spp: This group includes Clarias gariepinus (Burchell) and C. liocephalus
271 (Boulenger) which is quite rare in the lake and has been observed only in open waters. C.
272 gariepinus has been considered as the key species for this group due to its predominance and
273 well-studied ecological characteristics (B. Kaningini, UERAH, pers. comm.). Ulyel (1991)
274 considered this species as a benthic polyphage, feeding on insects, crustaceans and fishes.
275
276 2°) Raiamas moorei: This species, formerly known as Barilius moorei (Boulenger), is the only
277 cyprinid inhabiting most areas of the lake. It feeds mainly on small bottom mollusks and insects,
278 as well as on small cichlids and clupeids. EE was set at 0.50 as this group is rarely targeted by
279 fishermen and seems to be exposed to a limited predation by Clarias spp and Haplochromis
280 vittatus (Boulenger).
281
282 3°) Haplochromis spp: A list of 17 species was proposed and observed by Snoeks (1994). For
283 the purpose of the present study, these species have been re-grouped based on their feeding
284 ecology (Snoeks, 1994; Ulyel et al., 1991; Fourniret et al., 1992; Kaningini et al., 1999; Fröese
285 and Pauly, 2006) into three groups:
286 • H. vittatus, a piscivore species which inhabits both the near shore and open waters.
287 • Benthos-feeding haplochromines are mostly H. gracilior (Boulenger), H. graueri
288 (Boulenger), H. astatodon (Regan), H. nigroides (Pellegrin) and H. paucidens
12 289 (Regan). They inhabit the littoral areas and feed mainly on small mollusks,
290 nematodes, insects and their larvae. The biological characteristics of H. graueri
291 were considered since this is the most abundant among the benthophages.
292 • Plankton-feeding haplochromines are dominated by H. kamiranzovu (Snoeks) which
293 was considered as the key species for this group.
294 Diet compositions of haplochromine groups were adapted from Ulyel et al. (1991) and
295 Fourniret et al. (1992).
296
297 4°) Limnothrissa miodon: This pelagic fish feeds mainly on zooplankton (Copepods) during its
298 juvenile stage and may consume insects and small-sized fishes (de Iongh et al., 1983; Kaningini,
299 1995; Isumbisho et al., 2004). According to Pearce (1995), it is capable of adapting its diet
300 preferences according to the local conditions (i.e., food availability), as this species is not
301 specialized with regards to its preys. Demographic studies have been carried out by de Iongh et
302 al. (1995) and Kaningini (1995). According to de Iongh et al. (1995), three length classes can be
303 identified for this species based on condition factor changes. For this study, however, only two
304 length classes were considered based on the length at first maturity (50% of the gonads
305 maturing) to separate the pelagic adults (>8.0 cm total length) from the sub littoral juveniles (<
306 8.0 cm total length). This repartition takes into account the difference of their spatial distribution
307 (Spliethoff et al., 1983; Lambœuf, 1991), spawning-related migrations (Marshall, 1991) and
308 ontogenetic diet variations (de Iongh et al., 1983). Several studies on feeding patterns have also
309 indicated occurrence of cannibalism (Spliethoff et al., 1983; Isumbisho et al., 2004).
310
311 It should be noted that the adults have a low P/B ratio (1.75 yr-1) compared to juveniles
312 (6.69 yr-1), a feature which was already noted when juveniles and adults are separated in an
13 313 Ecopath model such as for Lates niloticus (Linnaeus) in Lake Victoria (Villanueva and Moreau,
314 2001). It is due to higher exploitation and predation on the latter, which are also integrated into
315 the adult pool. As juveniles sardines prey mainly on zooplankton and small benthos, their Q/B
316 values measured using MAXIMS are also higher (35.8 yr-1) than those of adults (19.20 yr-1)
317 which feed on zooplankton but also on small fish. In addition, small fish consume much more
318 food relatively to their size than larger ones (Pauly and Palomares, 1987).
319
320 5°) Barbus spp inhabit near-shore area. This group includes Barbus kerstenii (Peters), B
321 pellegrini (Poll), B. pleurogramma (Boulenger) and B. altianalis (Boulenger). This group feeds
322 basically on microphytes, insects and other benthic organisms, and on small littoral fishes (Ulyel,
323 1991; Kaningini et al., 1999; Fröese and Pauly, 2006). This group is one of the least consumed
324 and exploited and least predated upon in the lake; EE was, therefore, admitted to be 0.50.
325
326 6°) Oreochromis niloticus eduardianus (Boulenger) is an endemic, microphage fish inhabiting
327 the littoral areas. It is the most abundant, native cichlid species in the lake due to its high
328 fecundity (Trewavas, 1983).
329
330 7°) Other Tilapiine fish include two introduced cichlid species, O. macrochir and T. rendalli),
331 with the latter being more abundant possibly due to its efficient reproduction and feeding
332 plasticity (Trewavas, 1983). In its original environment, T. rendalli is, however, regarded as a
333 macrophyte-feeding species (Fröese and Pauly, 2006). Both inhabit the littoral areas of the lake
334 and feed on macrophytoplankton and other organic material (Ulyel, 1991; Kaningini et al.,
335 1999).
336
14 337 Results
338
339 The basic input for each group and the parameters computed by the model are presented
340 in Table 2 and 4, while the relative diet compositions are given in Table 3 whereas Figure 2
341 summarizes the main flows within this ecosystem. Total estimated fish biomass is low, 3.705
- 342 t.km ², compared to other African inland waters (Christensen and Pauly, 1993). The resulting
- 343 biomass for benthos looks low (3.676 t.km ²) for the whole lake, though concentration along the
- 344 littoral zones yields 36.76 t.km ². This is in agreement with the importance of this group in the
345 diets of several fish groups. A high abundance of zoobenthic organisms has already been noted
346 in shallow areas of Lake George (Moreau et al., 1993) or Lake Ihéma (Mavuti et al., 1996).
347
348 Highest TLs were estimated for Clarias spp, R. moorei and H. vittatus (TL >3.3) due to
349 their carnivorous feeding ecology (Table 3). Most groups belonging to TL3 and more are
350 predatory carnivores.
351
352 Ecotrophic efficiencies
353 A low EE value of 0.079 has been calculated for the detritus as most of it sinks to the
354 bottom in the deepest parts of the lake. Phytoplankton has a higher EE of 0.633, indicating that
355 this group is the base food source in the lake even if it seems that it is not fully utilized by
356 organisms of higher TLs, at least in this area of the lake. This may be attributed to the limited
357 quantity of fish basically consuming this group. A high value of EE is noted as well for
358 zooplankton (0.764). Carnivorous zooplankton (Copepods) partly feed on the herbivorous
359 zooplankton (mainly Cladocerans and Rotifers), as noted in the feeding matrix (Table 2)
360 although they consume also phytoplankton. Ecotrophic efficiencies of fish groups are variable.
15 361 For the two groups of Tilapiine fish, EE is quite low (EE = 0.479 and 0.233), suggesting a very
362 limited exploitation and predation in the lake. The maximum EE value (0.917) is recorded for L.
363 miodon juveniles, as we expected because these fish are exploited and predated.
364
365 The gross efficiencies
366 The P/Q ratios (Table 4) are low for R. moorei and adult L. miodon. This might be due to
367 the low density of their prey, particularly the zooplankton, and the necessity for these fish to use
368 more energy for hunting their prey, which are available only at low densities. It should be noted
-3 369 that the density of zooplankton per volume basis is very low: 0.0914 g.m fresh weight
370 (Isumbisho et al., 2006). The low P/Q ratios obtained for the 2 groups of Tilapiine fish (0.040
371 and 0.045) are in agreement with the low quality of their preferred preys which are principally
372 phytoplankton and decaying organic material. A high P/Q ratio is estimated for zoophagous
373 haplochromines (0.202) due to their carnivorous feeding habits. The maximum value (0.220) was
374 obtained for the juveniles L. miodon and this is in relationship with their small size. This value is
375 higher than for adults, which is in agreement with the basis of the method of computation of Q/B
376 implemented by Pauly and Palomares (1987). Ichthyophagous fishes (Clarias spp and H.
377 vittatus) have surprisingly low P/Q values. It might come from the scarcity of their possible prey
378 in terms of biomass per volume unit.
379
380 Omnivory indices and diet overlap
381 The omnivory index (OI) of each group is presented in table 4. In general, high OI values
382 are observed in high TLs, which indicate more complexity in this part of the food web. Highest
383 OI are observed for three predators: Clarias spp, Barbus spp and H. vittatus (0.282, 0.349 and
384 0.339, respectively), and is related to their large feeding spectrum and distribution in the lake.
16 385 These observations concord with the indication of Lindeman (1942) that prey tend to be more
386 specialized than their predators.
387
388 Adults of L. miodon have a lower OI (0.153), than their juveniles (0.218) due to the
389 latter’s feeding flexibility (Isumbisho et al., 2004). Accordingly, our results indicate that, in
390 habitats where it is already acclimated, L. miodon adjusts its trophic behavior to the availability
391 of aquatic macro invertebrates. This has already been noted by Marshall (1995), Kaningini
392 (1995) and recently by Isumbisho et al. (2004). Compared to H. vittatus, Haplochromines groups
393 4 and 5 have a lower OI (0.155 and 0.179, respectively), which suggests a higher specialization.
394
395 OI of O. niloticus is zero as this fish consumes only preys from the first trophic level,
396 mainly phytoplankton (Table 3). Preference for phytoplankton of this species has been observed
397 in other tropical lakes (Tadesse, 1999; Lu et al., 2006). Other Tilapiines, on the other hand, have
398 a higher OI due to their trophic plasticity, particularly for T. rendalli, that enables dietary shifts
399 from plant or detrital material to animal material (Ulyel 1991; Kaningini et al., 1999).
400
401 High values of individual OI for groups sharing the same type of food can be associated
402 with estimates of niche overlaps. L. miodon juveniles (group 7) have a low overlap as predator or
403 prey for most groups, except for the adults (group 6) which show high overlap in prey (Fig. 3).
404 Groups 9 (O. niloticus) and 10 (Other Tilapiines) show the highest overlap which suggests high
405 competition for similar resources. Groups 2 (R. moorei) and 7 have the lowest overlap which
406 expresses divergent preferences in terms of resources consumed (Fig. 3).
407
17 408 The SOI for the southern part of Lake Kivu is low, 0.150 with a connectance index (CI)
409 of 0.396 (Table 4). This CI value is slightly higher than the theoretical value (0.317) computed
410 using the regression model of Christensen and Pauly (1993). Both values can imply that most
411 functional groups exhibit a certain degree of diet specialization. This indicates as well the co-
412 existence of weak and strong interactions observed among groups as expressed by the various EE
413 values (Table 2). According to Quince et al. (2002), this is common in food webs of especially
414 competitive communities which might be the case here. Moreover, McCann (2000) indicated that
415 recurrent food-web structures, with omnivory and apparent competition, can enhance ecosystem
416 stability if the distribution of consumer–resource interaction strengths is skewed towards weak
417 interaction strengths and McCann referred to as the “weak-interaction effect” which contributes
418 to community-level stability.
419
420 Biomass flux and transfer efficiency
421 Trophic aggregation revealed that transfer efficiency from TL1 (phytoplankton and
422 detritus groups combined) to higher TLs is about 8.4%. This indicates that this resource may not
423 be fully exploited due to the presence of herbivores in the littoral area, which makes up only 10%
424 of the lake, leading to increased unconsumed nutrient accumulation (Table 5). The average
425 transfer efficiency is at any TL is less than 10% (Table 5). This is higher than that observed in
426 other ecosystems such as Lake Navaisha (Moreau et al., 2001). Most of the fish biomass and
427 ecological production take place at TL3 or more, as summarized in table 5.
428
429 The ecosystem is phytoplankton-based as 61% of the total flow originating from TL1
430 comes from primary producers while only 39% originates from detritus (Table 5), a feature of
431 relevance in a deep-water body (Christensen and Pauly, 1993). Most primary production is
18 432 consumed by zooplankton and juveniles L. miodon. Detritus is consumed only by benthic fauna
433 (group 11) and, to a certain degree, by some fish groups: Barbus spp., O. niloticus and other
434 Tilapiines (Table 3). The elevated proportion of primary production flowing back to detritus
435 (about 65 % of the total) is the result of increasing algal biomass surplus unconsumed, especially
436 in the open waters of the lake.
437
438 A high ratio between production and respiration (3.92) is noted. Most likely, a limited
439 quantity of organic matter is imported by inflowing rivers. An important part of the production of
440 several groups is not utilized (EE is low) and is therefore lost as incorporated into the sediments
441 on the bottom of the lake which has a deep anoxic hypolimnion. This might explain this
442 unusually high value of the production/respiration value.
443
444 Fish productivity is linked to primary production by many intermediate trophic links. The
445 primary production required (PPR) in order to support the fishery is 15.2% of the total primary
446 production (Table 6) which is low compared to an average value suggested by Pauly and
447 Christensen (1995) for tropical lakes and rivers (23.6 %). When expressed relative to the total
448 flow from TL1, the PPR (primary production required) used in Ecopath corresponds to the
449 ecological footprint (EF). For the catch the expression is EF = PPR/PP*C where PP is the total
450 flow from TL1, and C the catch, will give the size of the area in km², assuming the unit is,
451 needed to sustain a catch of 1 ton for the given resource. As a consequence, a low ecological
452 footprint (Folke and Kautsky, 1996) of the fishery (0.04 km²) is observed, similar to those
453 observed by Villanueva et al. (2006) in two West African lagoons.
454
19 455 The GEF is quite low (0.0015) compared to what was observed for Lake George (Moreau
456 et al., 1993), Lake Ihéma (Mavuti et al., 1996) and Lake Victoria (Villanueva and Moreau, 2001)
457 or other tropical inland water bodies (Christensen and Pauly, 1993). The mean TL of the fishery
458 is 2.9 (Table 6) as it targets mostly L. miodon.
459
460 Model predictions of the effects of environment changes
461
462 The mixed trophic impact (MTI) routine of Ecopath (Ulanowicz and Puccia, 1990) shows
463 the direct and indirect influences of abundance variations of any species group on all other
464 groups considered (Fig. 4). An initial condition that should be considered for this routine is that
465 diet composition of each functional group does not change, despite possible variations in
466 abundance of their various preys. An increased abundance of fish groups of high TLs (about 3 or
467 more) would have various levels of negative impacts on other groups. This is particularly the
468 case for Clarias spp (group 1) and L. miodon. An increasing abundance of non fish groups would
469 generate a positive impact on most groups including fish groups. The impact of zooplankton
470 biomass variations would be less important compared to the phytoplankton group. The extent of
471 bottom-up control is elevated, as an increase in abundance of phytoplankton would have a strong
472 positive effect on all higher TLs (Fig. 4), especially on the herbivores (groups 5, 9, 10 and 12).
473
474 A top-down trophic cascade effect (Pace et al., 1999; Persson, 1999) on phytoplankton
475 biomass is also observed in the MTI simulation (Fig. 4). The primary productivity is increased
476 due to the increase in planktivorous fish (TL3) that regulates herbivores, which in turn prey on
477 phytoplankton. Hence, this may lead to a build-up of nonutilized phytoplankton. This was also
478 experimentally observed by Lynch and Shapiro (1981).
20 479
480 Figure 5 shows the susceptibility of some fish to human exploitation than others. Fishing
481 gears employed capture a specific species in the lake (Hanek et al., 1991). An increase of 10% in
482 the fishing effort shows a substantial increase in catch of target species, i.e. Clarias spp.,
483 Haplochromines, adult and juvenile L. miodon. Positive impacts on groups at lower TLs is a
484 consequence of lower predation pressure when stock of fish predators decline due to increased
485 fishing. An increasing fishing effort with beach seine and longline would have a slight negative
486 impact on zooplankton. Both gears do not target L. miodon juveniles which are the principal
487 predator of this group.
488
489 Discussion
490
491 Successful colonization of L. miodon in Lake Kivu has been attributed to the absence of
492 other pelagic planktivores (de Iongh et al., 1995; Marshall, 1995; Munyandorero and Mwape,
493 2003) and low diversity of native species (Johannesson and Lambœuf, 1989). These combined
494 with the relative stability of environmental factors suitable for its growth, provided opportunities
495 for successful colonization of this sardine (Marshall, 1991; 1995) and other Cichlids.
496 Acclimatization of stocked species has been also observed in other African lakes such as Lakes
497 Kyoga (Ogutu-Ohwayo, 1990), Kariba (Karenge and Kolding, 1995), Navaisha (Muchiri et al.,
498 1995; Moreau et al., 2001), Nabugabo (Chapman et al. 1996) and Tana (de Graaf et al., 2000) as
499 well as in other ecosystems in the world (Vitousek et al., 1997; Wilcove et al., 1998; Latini and
500 Petrere Jr., 2004). Environmental condition modifications have already been cited as a major
501 factor in enhancing long-term success and dominance of exotics species in several ecosystems
502 (Muchiri et al., 1995; Smith et al., 2000; Dudgeon et al., 2006).
21 503
504 The ability of L. miodon to coexist with other zooplanktivores may be essentially due to
505 spatial heterogeneity, thus, an absence of co-adapted competitors. Competition between similar
506 functional groups may also alleviate indirect effects of predators on ecosystem processes and
507 exhibits functional redundancy in ecosystems (Lawton and Brown, 1993; Loreau et al., 2001;
508 Raffaelli et al., 2002; Stachowicz et al., 2002). Similar observations in other ecosystems were
509 made in Lake Kariba (Karenge and Kolding, 1995), Parakrama Samudra reservoir (Moreau et
510 al., 2001) and in the Great Lakes (Mills et al., 1993). Coexistence seems possible based on a
511 competitive exclusion principle (Richards et al., 2000) by limiting competition through space
512 budgeting (Isumbisho et al., 2003) between or within-guild species. Some haplochromines
513 inhabit mostly the inshore zone while L. miodon occupies essentially the open waters. The
514 significance of spatial heterogeneity in favoring increase of species abundance has also been
515 observed by Le Pape et al. (2003) in the Bay of Biscay.
516
517 Similar to L. miodon, naturalized tilapias still remain at limited levels without major
518 impacts on the indigenous Nile Tilapia. Coexistence of both endemic and alien tilapias may be
519 due to niche partitioning aside from the broad tolerance of tilapias against environmental
520 changes (Murichi et al., 1995; Iwama et al., 1997; Khallaf et al., 2003). In Lake Kivu, these
521 tilapias inhabit essentially the shallow waters (>10 m deep). The endemic O. niloticus
522 eduardianus, however, is abundant in rocky bottoms while introduced Tilapiine fishes (O.
523 macrochir and, especially, T. rendalli) colonize better the muddy littoral zones (Trewavas,
524 1983). Spatial segregation limits competition for food and nursery sites similarly observed in
525 Lakes Victoria and Kyoga (Twongo, 1995). O. niloticus and O. macrochir are both microphages
526 which may explain the elevated prey-predator overlap (Fig. 3). Resistance of O. niloticus
22 527 eduardianus may be mediated by its opportunistic behavior despite dietary overlap with O.
528 macrochir. Broadening of diet spectrum can increase a species tolerance to stress (Murichi et al.,
529 1995; Sax and Brown, 2000; Wanink and Witte, 2000; Villanueva et al., 2006). Dietary shifts of
530 O. niloticus are similarly observed in Lake Victoria (Njiru et al., 2004).
531
532 This, however, may not apply to other species of lower environmental tolerance in the
533 lake. Low omnivory indices were observed for some groups, indicating a less diversified diet.
534 This is not the case for the sub-littoral inhabiting cichlids which contribute to the efficient
535 utilization of some resources, i.e. here the primary producers. Despite the low contribution of the
536 detritus group, it is still utilized as a buffering agent in case of resource limitation. Similar
537 observations were indicated by in Lake Navaisha (Munichi et al., 1995; Mavuti et al., 1996) and
538 in some West African lagoons (Villanueva et al., 2006).
539
540 Both the predation-based regulation of the lower TLs and the resource-based regulation of
541 the upper TLs are present in the ecosystem studied. The strength of both bottom-up and top-down
542 controls determine system diversity though their relative importance and intensity which are
543 based on the structure and functioning of groups among TLs (Herendeen, 2004). Differences in
544 migration patterns and food availability may influence predation rates of predators, such as C.
545 gariepinus, H. vittatus and L. miodon (Isumbisho et al., 2004). This is similar to observations of
546 Bruton (1979) and Huddart (1994) in other African lakes.
547
548 The MTI analysis (Fig. 4), nonetheless, demonstrated the importance of bottom-up forces
549 through the strong potential influence of any phytoplankton abundance variation on the whole
550 food web. In ecological theory, bottom-up forces would dominate the ecosystem process (Platts
23 551 and Ulanowicz, 1985; Dyer and Letourneau, 2003). According to Proulx et al. (1996),
552 production can also be modified through algal community structure modification as a function of
553 variations in size distribution (Perin et al., 1996) or through predation-mediated modifications in
554 plankton community structures (Hansson and Carpenter, 1993; Dyer and Letourneau, 2003).
555
556 Isumbisho et al. (2004) observed variations on zooplankton community abundance and
557 distribution which is mainly due to predation by L. miodon juveniles and partly, in relation, to
558 fishery practices. Predation-related variations within functional groups may lead to
559 compensatory population increases by less vulnerable species in response to predation on
560 vulnerable species (Steiner, 2001). Predation and competition affect aquatic communities
561 indirectly by causing behavioral changes in prey species (Nyström et al., 2001; Steiner, 2001).
562 Predation is an important part of ecosystem functioning though an increase in aggressive
563 invaders or top predators can equally multiply effects of feeding interactions and contribute to
564 reorganization of ecosystem structure indirectly (Fulton et al., 2003; Didham et al., 2005).
565 Changes in zooplankton community structure have also been observed in Lake Donghu (China)
566 by Yang et al. (2005) as a consequence of increased predation pressure of zooplanktivores. In
567 Lake Victoria such phenomenon is attributed to eutrophication (Wanink et al., 2002).
568
569 The introduction of L. miodon and tilapias in the lake has surely improved energy transfer
570 efficiencies in the lake, especially in the pelagic zone. Significant changes in functional roles at
571 individual (i.e., changes in behavior and habitat use) and population (i.e. change in the
572 abundance and distribution) levels were observed in the lake. Species change has also
573 contributed to an entirely new configuration of the fisheries (Van den Bossche and Bernacsek,
574 1990; Preikshot et al., 1998). Compared to other ecopath-modeled deep lakes in Africa (see
24 575 Villanueva and Moreau, 2001), however, there are many functional groups in Lake Kivu with
576 low EE values which imply that lower TLs are not saturated and predator abundance variations
577 can create vacant niches leaving unexploited resources in lower TLs. Functional consequences of
578 low biodiversity and species specific traits (i.e., size, trophic role, rarity, distribution and degree
579 of specialization) or combined effects of both may explain inefficiencies of energy transfers in
580 this lake. Lower energy transfer efficiencies in high TLs have been observed in other poorly
581 diverse systems (Loreau et al., 2001; Raffaelli et al., 2002; Stachowicz et al., 2002).
582
583 Fish stocking is an important aspect of fishery management but comes with considerable
584 risks. Invasive species can redefine an ecosystem by converting diverse communities into mono
585 specific ones as introduced species often become invasive and may lead to native species
586 extinctions (McKinney and Lockwood, 1999; Simberloff, 2000; Lodge, 2001; Rosenzweig,
587 2001; Davis and Thompson, 2000; de Graaf et al., 2000; Mack et al., 2000; Pimentel et al., 2001;
588 Dudgeon et al. 2006). In African freshwater systems, an ominous example is the loss of
589 biodiversity in Lake Victoria following the Nile perch introduction (Kudhongania and
590 Chitamwebwa, 1995; Pitcher and Hart, 1995; Preikshot et al., 1998). Biodiversity in Lake Kivu,
591 however, is low and fish were stocked to boost up the lake’s biodiversity and productivity.
592
593 The role of biodiversity has been hypothesized as insurance to ecosystem functioning in
594 case of modifications (Loreau et al., 2001). Biodiversity after introduction is affected in two
595 manners either by global homogenization of regional biota or by affecting native species
596 functions (Levine et al. 2003; Didham et al., 2005; Korniss and Caraco, 2005; Puth and Post,
597 2005; Olden and Rooney, 2006). Predators and resources manipulations can cause direct changes
598 of diversity at one TL, which in turn, affect diversity of other TLs. Indirect biodiversity effects of
25 599 varying resources and consumers, on the other hand, are supposed to be stronger in aquatic
600 ecosystems compared to terrestrial systems and detritus based food webs (Dyer and Letourneau,
601 2003). Intraspecific food resource limitation had been observed on adults of L. miodon which
602 had evolved from a mere pelagic planktivore to a piscivore, preying on its juveniles (Lowe-
603 McConnell, 1993). The sequestration of space and nutrients by adults may limit resources that
604 eventually provoked such cannibalistic behavior similar to observations of Mandima (1999) on
605 L. miodon in Lake Kariba and by Bundy and Pitcher (1995) on Nile Perch in Lake Victoria.
606
607 In Lake Kivu, however, fish introductions showed no detrimental changes at both the
608 biodiversity and ecological levels of the fish community (Marshall, 1995; Ogutu-Ohwayo et al.,
609 1997). It can be suggested that, under non-limiting food conditions, L. miodon which is
610 occupying mainly the open waters of the lake do not have a particular deleterious effects on other
611 indigenous species, i.e. provoking mass extinction of the latter. It actually represents the lake’s
612 most important stock in terms of biomass and production (Hanek et al., 1991). It has also
613 stimulated the commercial fishery with an estimated potential annual yield of 55 kg.ha-1 in the
614 early 90s (Johannesson and Lambœuf, 1989; van den Bossche and Bernacsek, 1990; Hanek et
615 al., 1991). At present, introduced fish species in this lake comprise 85% of the catch (B.
616 Kaningini, UERHA, pers. data). A similar phenomenon observed in Lake Kariba (Coulter et al.,
617 1984; Marshall and Mubamba, 1993). Despite the increasing intensity of the fishing activity on
618 exotics, they seem resilient to human exploitation. Tilapias are known to be highly resilient to
619 intensive fisheries due to their unspecialized ecological flexibility.
620
621 Based on the classification of colonizers discussed by Davis and Thompson (2000), L.
622 miodon is a type 2 colonizer due to its eventual key role function in the ecosystem (Spliethoff et
26 623 al., 1983; Isumbisho et al., 2006), especially in the fisheries (van den Bossche and Bernacsek,
624 1990; Marshall and Mabamba, 1993). In most cases, exotic fishes stocked have evolved
625 pervasive in other lakes not only in Africa (Mills et al., 1993; Ogutu-Ohwayo et al. 1997; de
626 Graaf et al., 2000) but worldwide (Villanueva and Moreau, 2001; Dudgeon et al., 2006). Once
627 the biodiversity is altered, the ecosystem is transformed into new configurations often
628 detrimental to human welfare (Ruesink et al., 1995; McCann et al., 2001; Pimentel et al., 2005;
629 Dudgeon et al., 2006; Lovett et al., 2006). In Lake Victoria, for example, the Nile Tilapia has
630 evolved as a “keystone species” though its introduction led to the alteration of ecosystem
631 function, biotic interactions and biotic homogenization (Moreau, 1995; Bundy and Pitcher, 1995,
632 Villanueva and Moreau 2001).
633
634 Conclusion
635
636 For this study, we have tried to quantify the impact of invasive species on ecosystem
637 process and functioning. Based on this study, energy fluxes from primary producers in this
638 ecosystem are largely phytoplankton-based because of the importance of zooplankton and
639 abundance of zooplankton consumers. The fragility of the ecosystem to environmental changes
640 lies, therefore, in the production of the zooplankton group (TL 2), which evidently links the
641 transfer of primary production to higher TLs, and in the paucity of species diversity. The existing
642 competitions for food among some groups can be limited mostly by the specialization of some of
643 them to narrow spatial distributions.
644
645 Broad-scale or ecosystem-level approach is recognized as crucial in describing and
646 understanding the trophic structure in Lake Kivu and the importance of the introduced species. It
27 647 is a requirement to elucidate and, eventually, predict possible impact of exotic species on natural
648 food webs. Fish introductions in Lake Kivu is interesting in the sense that the introduction of
649 exotic species have increased energy flux transfers between TLs for what may have been initially
650 a ‘dead zone’ before the sardine colonization. Exotics play key role functions in the ecosystem
651 (Spliethoff et al., 1983; Isumbisho et al., 2006), especially in the fisheries (van den Bossche and
652 Bernacsek, 1990; Marshall and Mabamba, 1993). It is possibly one of the rare occasions where
653 introduction of exotics can be acknowledged as positive in terms stimulating commercial
654 fisheries (Johannensson and Lambœuf, 1989; van den Bossche and Bernacsek, 1990; Hanek et
655 al., 1991) and improving fish biodiversity with minor consequences on trophic structure and
656 functioning.
657
658 Acknowledgements
659
660 The authors wish to thank M. Masilya, M. Waasalo and H. Sarmento for useful
661 information and comments. M. Isumbisho acknowledges IFS (International Foundation for
662 Science) for financial support that allowed the completion of this work. The bulk of the field
663 work was supported by the “Agence Universitaire de la Francophonie” (AUF, Programme de
664 Coopération Scientifique Inter-Universitaire). Finally, we are indebted to the two anonymous
665 referees for valuable comments on the earlier version of the manuscript.
666
667 References
668
669 Allen, K.R. 1971. Relation between production and biomass. Journal of the Fisheries
670 Research Board of Canada, 28: 1573-1581.
28 671
672 Allen, E.A. and Humble, L.M. 2002. Nonindigenous species introductions: A threat to Canada's
673 forests and forest economy. Canadian Journal of Plant Pathology, 24:103-10.
674
675 Arim, M., Abades, S.R., Neill, P.E., Lima, M. and Marquet, P.A. 2006. Spread dynamics of
676 invasive species. Proceedings of the National Academy of Sciences, 103(2): 374-378.
677
678 Bruton, M.N. 1979. The role of diel inshore movements by Clarias gariepinus (Pisces:
679 Clariidae) for the capture of fish prey. Transactions of the Zoological Society of London, 35:
680 115-138.
681
682 Bundy, A. and Pitcher, T.J. 1995. An analysis of species changes in Lake Victoria: did the Nile
683 perch act alone?, pp. 111-135. In: T. Pitcher and P.J.B.Hart (Eds.), The impact of Species
684 Changes in African Lakes. Fish and Fisheries Series, 18, Chapman and Hall, London.
685
686 CapBlanc, J. and Dauta, A. 1999. Photosynthesis simulator ver 2.1. Université Paul Sabatier and
687 C.N.R.S., Toulouse.
688
689 Chapman, L.J., Chapman, C.A., Ogutu-Ohwayo, R., Chandler, M., Kaufman, L. and Keiter, A.E.
690 1996. Refugia for endangered species from an introduced predator in Lake Nabugabo,
691 Uganda. Conservation Biology, 10: 554-561.
692
693 Christensen, V. and Pauly, D. (Eds.) 1993. Trophic Models of Aquatic Ecosystems. ICLARM
694 Conference Proceedings, 26: 390 pp.
29 695
696 Christensen, V. 1995. Ecosystem maturity - towards quantification. Ecological Modelling, 77:
697 3-32.
698
699 Christensen, V. and Walters, C. 2004a. Ecopath with Ecosim: methods, capabilities and
700 limitations. Ecological Modelling, 172: 109-139.
701
702 Christensen, V. and Walters, C. 2004b. Trade-offs in ecosystem-scale optimization of fisheries
703 management policies. Bulletin of Marine Science, 74 (3): 549-562.
704
705 Christensen, V., Walters, C. and Pauly, D. 2005. Ecopath with Ecosim: a User’s Guide. Fisheries
706 Centre Research Reports, 12 (4): 154 pp.
707
708 Coulter, G.W., Allanson, B.R., Bruton, M.N., Greenwood, P.H., Hart, R.C., Jackson, P.B.N. and
709 Ribbink, A.J. 1984. Unique qualities and special problems of the African Great Lakes.
710 Environmental Biology of Fishes, 17(3): 161-183.
711
712 Craig, J.F. 1992. Human-induced changes in the composition of fish communities in the African
713 Great Lakes. Review in Fish Biology and Fisheries, 2: 93-124.
714
715 Crawley, M.J. 1987. What makes a community invasible?, pp. 429-453 In: A.J. Gray, M.J.
716 Crawley and P.J. Edwards (Eds.), Colonization, Succession and Stability. Blackwell Scientific
717 Publications, Oxford.
718
30 719 Davis, M.A. and Thompson, K. 2000. Eight ways to be a colonizer, two ways to be an invader: a
720 proposed nomenclature scheme for invasion ecology. Bulletin of the Ecological Society of
721 America, 81: 226–230.
722
723 de Graaf, M., Dejen, E., Sibbing, F.A. and Osse, J.W.M. 2000b. Barbus tanapelagius, a new
724 species from Lake Tana (Ethiopia): its morphology and ecology. Environmental Biology of
725 Fishes, 59: 1–9.
726
727 de Iongh, H.H., Splithoff, P.C. and Frank, V.G. 1983. Feeding habits of the clupeid Limnothrissa
728 miodon (Boulenger) in Lake Kivu. Hydrobiologia, 102: 113-122.
729
730 de Iongh, H.H., Spliethoff, P.C. and Roest, F. 1995. The impact of an introduction of sardine into
731 Lake Kivu, pp. 278-297. In: T. Pitcher and P.J.B. Hart (Eds.), The impact of Species Changes in
732 African Lakes. Fish and Fisheries Series 18, Chapman and Hall, London.
733
734 De Merona, B. 1983. Modèle de croissance des poissons d’eau douce d’Afrique. Revue
735 d’Hydrobiologie Tropicale, 19 : 1-12.
736
737 Descy, J.-P. and Fourniret, Y. 1991. Quelques données récentes sur la production planctonique
738 du lac Kivu (Rwanda), pp.147-149. In P. Vernet (Ed.), Hommage à F.A. FOREL, Proceedings of
739 the 3rd CILEF, Morges, Switzerland.
740
741 De Vos, L., Snoeks, J. and Thys van den Audenaerde, D.K. 2001. An annotated checklist of the
742 fishes of Rwanda (East-central Africa) with historical data on introductions of commercially
31 743 important species. Journal of East African Natural History, 90: 41-68.
744
745 Didham, R.K., Tylianakis, J.M., Hutchison, M.A., Ewers, R.M. and Gemmell, N.J. 2005. Are
746 invasive species the drivers of ecological change? Trends in Ecology and Evolution, 20 (9):
747 470-474.
748
749 Dudgeon, D., Arthington, A.H., Gessner, M.O., Kawabata, Z.-I., Knowler, D.J., Lévêque, C.,
750 Naiman, R.J., Prieur-Richard, A.-H., Soto, D., Stiassny, M.L. and Sullivan, C.A. 2006.
751 Freshwater biodiversity: importance, threats, status and conservation challenges. Biology
752 Review, 81: 163-182.
753
754 Dyer, L.A. and Letourneau, D. 2003. Top-down and bottom-up diversity cascades in detrital vs.
755 living food webs. Ecology Letters, 6: 60-68.
756
757 Folke, C. and Kautsky, N. 1996. The ecological footprint concept for sustainable seafood
758 production. Proceedings of the International Conference on “Ecosystem Management for
759 Sustainable Marine Fisheries” 19-23 February 1996. Monterey, California.
760
761 Fourniret, Y., Plisnier, P.D. and Micha, J.-C. 1992. Régime alimentaire de quatre espèces du
762 genre Haplochromis (Téléostéi, Cichlidae) du lac Ihéma (Rwanda) Annales de Limnologie, 28
763 (1): 57-69.
764
765 Fröese, R. and Binohlan, C. 2000. Empirical relationships to estimate asymptotic length, length
766 at first maturity and yield at maximum yield per recruitment in fishes, with a simple method to
32 767 evaluate length frequency data. Journal of Fish Biology, 56: 758-773.
768
769 Fröese, R. and Pauly, D. (Eds.) 2006. Fishbase. http://www.fishbase.org.
770
771 Fulton, E.A., Smith, A.D.M. and Johnson, C.R. 2003. Mortality and predation in ecosystem
772 models: is it important how these are expressed? Ecological Modelling, 169: 157-178.
773
774 Gayanilo, F.C. Jr., Sparre, P. and Pauly. D. (Eds.) 2002. The FAO-ICLARM Stock Assessment
775 Tools II (FiSAT II Ver. 1.0) FAO http://www.fao.org/fi/statist/fisoft/fisat/.
776
777 Gottlieb, S.J. and Schweighofer, M.E. 1996. Oysters and the Chesapeake Bay Ecosystem: A case
778 for exotic species introduction to improve environmental quality? Estuaries, 19 (3): 639-650.
779
780 Guo, Q., Qian, H., Ricklefs, R.E. and Xi, W. 2006. Distributions of exotic plants in eastern Asia
781 and North America. Ecology Letters, 9 (7): 827-834.
782
783 Gurevitch, J. and Padilla, D.K. 2004. Are invasive species a major cause of extinctions? Trends
784 in Ecology and Evolution, 19 (9): 470-474.
785
786 Hanek, G., Leendertse, K. and Farhani, B. 1991. Socio-Economic Investigations of Lake Kivu
787 Fisheries. UNDP/FAO Regional Project for Inland Fisheries Planning (IFIP), RAF/87/099-
788 TD/23/91 (En): 55 p.
789
33 790 Hansson, L. and Carpenter, S.R. 1993. Relative importance of nutrient availability and food
791 chain size for community composition in phytoplankton. Oïkos, 67: 257-263.
792
793 Harris, Z. 1968. Mathematical Structures of Language, Wiley, New York. Var. pag.
794
795 Hauser, L., Carvalho, G.R. and Pitcher, T.J. 1995. Morphological and genetic differentiation of
796 the African clupeid Limnothrissa miodon 34 years after its introduction to Lake Kivu. Journal of
797 Fish Biology, 47: 127-144.
798
799 Herendeen, R.A. 2004. Bottom-up and top-down effects in food chains depend on functional
800 dependence: an explicit framework. Ecological Modelling, 171: 21–33.
801
802 Hobbs, R.J. 1989. The nature and effects of disturbance relative to invasions, pp. 389-405. In: J.
803 Drake, F. di Castri, R. Groves, F. Kruger, H. Mooney, M. Rejmanek and M. Williamsom (Eds.),
804 Ecological invasions: A global perspective. Wiley, Chichester.
805
806 Huddart, R. 1994. Aspects of biology of the Lake Tanganyika sardine, Limnothrissa miodon
807 (Boulenger) in Lake Kariba. African Journal of Tropical Hydrobiology and Fisheries, 5: 31-
44. 808 44.
809
810 Irvine, K. and Waya, R. 1999. Spatial and temporal patterns of zooplankton standing biomass
811 and production in Lake Malawi. Hydrobiologia, 407: 191-205.
812
34 813 Irz, P., Argillier, C. and Proteau, J.-P. 2004. Contribution of native and non-native species to fish
814 communities in French reservoirs. Fisheries Management and Ecology, 11: 165-172.
815
816 Isumbisho, P., Kaningini, B., Descy, J-P. and Baras, E. 2004. Seasonal and diel variations in the
817 diet of the young stages of Limnothrissa miodon in Lake Kivu, Eastern Africa. Journal of
818 Tropical Ecology, 20: 1-11.
819
820 Isumbisho, P., Sarmento, H., Kaningini, B., Micha, J.-C. and Descy, J.P. 2006. Zooplankton of
821 Lake Kivu, East Africa: half a century after the Tanganyika sardine introduction. Journal of
822 Plankton Research, 28: 971-989.
823
824 Iwama, G.K., Takemura, A. and Takano, K. 1997. Oxygen consumption rates of tilapia in
825 freshwater, sea water and hypersaline sea water. Journal of Fish Biology, 51 (5): 886-894.
826
827 Jannasch, A.W. 1975. Methane oxidation in Lake Kivu (Central Africa). Limnology and
828 Oceanography, 20: 860-864.
829
830 Jarre, A., Palomares, M.L.D., Soriano, M., Sambilay, V. and Pauly, D. 1991. Some new
831 analytical and comparative methods for estimating the food consumption of fish. ICES Journal of
832 Marine Science, 193: 99-108.
833
834 Johannensson, K. and Lambœuf, M. 1989. Estimation acoustique de l’abondance de stock
835 d’Isambaza (Limnothrissa miodon) du lac Kivu, par une méthode basée sur le comptage des
836 échos de poissons, Gisenyi. RWA/87/012/TRAM/13, 64 pp.
837
35 838 Kaningini, B. 1995. Étude de la croissance, de la reproduction et de l’exploitation de
839 Limnothrissa miodon au lac Kivu, bassin de Bukavu (Zaïre). Thèse de doctorat, FUNDP,
840 Presses Universitaires de Namur, Belgium, 211p.
841
842 Kaningini, B., Micha, J.-C., Vandenhaute, J., Platteau, J.-P., Watongoka, H., Merlard, C.,
843 Wilondja, M.K. and Isumbisho, P. 1999. Pêche du Sambaza au filet maillant dans le lac Kivu.
844 Rapport final du Projet ONG/219/92/Zaïre. Presses Universitaires de Namur, Belgium.
845
846 Karenge, L. and Kolding, J. 1995. Inshore fish population and species changes in Lake Kariba,
847 Zimbabwe, pp. 245-275. In T. Pitcher and P.J.B. Hart (Eds.), The impact of Species Changes
848 in African Lakes. Fish and Fisheries Series 18 Chapman and Hall, London.
849
850 Kennedy, T.A., Naeem, S., Howe, K.M., Knops, J.M.H., Tilman, D. and Reich P. 2002.
851 Biodiversity as a barrier to ecological invasion. Nature, 417: 636-638.
852
853 Khallaf, E.A., Galal, M. and Authman, M. 2003. The biology of Oreochromis niloticus in a
854 polluted canal. Ecotoxicology, 12: 405-416.
855
856 Kolar, C.S. and Lodge, D.M. 2002. Ecological predictions and risk assessment for alien fishes in
857 North America. Science, 298:1233–1236.
858
859 Korniss, G. and Caraco, T. 2005. Spatial dynamics of invasion: The geometry of introduced
860 species. Journal of Theoretical Biology, 233: 137-150.
861
36 862 Kudhongania, A.W. and Chitamwebwa, D.B.R. 1995. Impact of environmental changes, species
863 introductions and ecological interactions on the fish stocks of Lake Victoria, pp. 19-32. In: T.
864 Pitcher and P.J.B. Hart (Eds.), The impact of Species Changes in African Lakes. Fish and
865 Fisheries Series 18, Chapman and Hall, London.
866
867 Lambœuf, M. 1991. Abondance et répartition du Limnothrissa miodon du lac Kivu, résultat des
868 prospections acoustiques d’avril 1989 à juin 1991. DOC/TR No. RWA/87/012/DOC/TR/46
869 (Fr) FAO, 41 p.
870
871 Latini, A.O. and Petrere Jr., M. 2004. Reduction of a native fish fauna by alien species: an
872 example from Brazilian freshwater tropical lakes. Fisheries Management and Ecology, 11: 71-
79. 873 79.
874
875 Lawton, J.H. and Brown, V.K. 1993. Redundancy in ecosystems, pp. 255-270. In: E.D. Schulze
876 and H.A. Mooney (Eds.), Biodiversity and ecosystem functions. Springer-Verlag.
877
878 Le Pape, O., Chauvet, F., Mahévas, S., Lazure, L., Guérault, G. and Désaunay, Y. 2003.
879 Quantitative description of habitat suitability for the juvenile common sole (Solea solea L.) and
880 contribution of different habitats to the adult population in the Bay of Biscay (France). Journal
881 of Sea Resources, 50: 139–149
882
883 Levine, J.M., Vilà, M., D’Antonio, C.M., Dukes, J.S., Grigulis, K. and Lavorel, S. 2003.
884 Mechanisms underlying the impact of exotic plant invasions. Philosophical Transactions of the
885 Royal Society of London, 270: 775–781.
37 886
887 Lindman, R.L. 1942.The trophic dynamic aspect of ecology. Ecology, 23: 399-418.
888
889 Lodge, D.M. 2001. Lakes, pp. 277–313. In: F.S. Chapin III, O.E. Sala and E. Huber-Sannwald
890 (Eds.), Global biodiversity in a changing environment: scenarios for the 21st century. Springer-
891 Verlag, New York.
892
893 Lodge, D.M., Stein, R.A., Brown, K.M., Covich, A.P., Brönmark, C., Garvey, J.E. and
894 Klosiewski, S.P. 1998. Predicting impact of freshwater exotic species on native biodiversity:
895 challenges in spatial scaling. Australian Journal of Ecology, 23: 53–67
896
897 Loreau, M., Naeem, S., Inchausti, P., Bengtsson, J., Grime, J.P., Hector, A., Hooper, D.U.,
898 Huston, M.A., Raffaelli, D., Schmidt, B., Tilman, D. and Wardle, D.A. 2001. Biodiversity and
899 Ecosystem Functioning: Current Knowledge and Future Challenges. Science, 294: 804-808.
900
901 Lovett, G.M., Canham, C.D., Arthur, M.A., Weathers, K.C. and Fitzhugh, R.D. 2006. Forest
902 ecosystem responses to exotic pests and pathogens in eastern North America. BioScience,
903 56 (5): 395-405
904
905 Lowe-McConnell, R.H. 1993. Fish Faunas of the African Great Lakes: Origins, Diversity, and
906 Vulnerability. Conservation Biology, 7 (3): 634-643.
907
908 Lu, K., Jin, C., Dong, S., Gu, B. and Bowen, S.H. 2006. Feeding and control of blue-green algal
909 blooms by tilapia (Oreochromis niloticus). Hydrobiologia, 568: 111-120.
38 910
911 Lynch, M. and Shapiro, J. 1981. Predation, enrichment and phytoplankton community structure.
912 Limnology and Oceanography, 26: 86-102.
913
914 MacDonald, I.A., Loope, L.L., Usher, M.B. and Hamann, O. 1989. Wildlife conservation and the
915 invasion of nature reserves by introduced species: a global perspective, pp. 215-255. In: J.A.
916 Drake, H.A. Mooney, F. di Castri, R.H. Groves, F.J. Kruger, M. Rejmdnek and M. Williamson
917 (Eds.), Biological invasions: A global perspective. John Wiley and Sons, Chichester.
918
919 Mack, R.N., Simberloff, D., Lonsdale, W.M., Evans, H., Clout, M. and Bazzaz, F.A. 2000.
920 Biotic invasions: Causes, epidemiology, global consequences, and control. Issues in Ecology, 5.
921
922 Mandima, J.J. 1999. Spatial and temporal variations in the food of the sardine Limnothrissa
923 miodon (Boulenger 1906) in Lake Kariba, Zimbabwe. Hydrobiologia, 407: 175-182.
924
925 Marshall, B.E. 1991. Seasonal and annual variations in the abundance of the clupeid
926 Limnothrissa miodon in Lake Kivu. Journal of Fish Biology, 39: 641-648.
927
928 Marshall, B.E. 1995. Why is Limnothrissa miodon such a successful introduced species and is
929 there anywhere else we should put it?, pp. 527-545. In: T.J. Pitcher & P.J.B. Hart (Eds.), The
930 impact of species changes in African Lakes. Fish and Fisheries Series 18, Chapman and Hall,
931 London.
932
39 933 Marshall, B.E. and Mubamba R. (Eds.) 1993. Biology, stock assessment and exploitation of
934 small pelagic fish species in the African Great Lakes Region. CIFA Occasional Paper 19, FAO,
935 Rome, 270 p.
936
937 Mavuti, K., Moreau, J., Munyandorero, J. and Pisnier, D. 1996. Analysis of trophic relationships
938 in two shallow equatorial lakes: Lake Naivasha (Kenya) and Lake Ihéma (Rwanda) using a
939 multispecifies trophic model. Hydrobiologia, 321: 89-100.
940
941 McCann, J. 2000. The diversity stability debate. Nature, 405: 228-233.
942
943 McKinney, M.L. and Lockwood, J.L., 1999. Biotic homogenization: A few winners replacing
944 many losers in the next mass extinction. Trends in Ecology and Evolution, 14: 450-453.
945
946 Mills, E.L., Leach, J.H., Carlton, J.T. and Secor, C.L. 1993. Exotic species in the Great Lakes:
947 A history of biotic crises and anthropogenic introductions. Journal of Great Lakes Research,
948 19(1): 1-54.
949
950 Moreau, J. 1995. Analysis of species changes in Lake Victoria using ECOPATH, a
951 multispecies trophic model, pp. 137-162. In T.J. Pitcher and P.J.B. Hart (Eds.), The impact of
952 species changes in African Lakes. Chapman Hall, London.
953
954 Moreau, J., Arrignon, J. and Jubb, R.A. 1988. Les introductions d’espèces étrangères dans
955 les eaux continentales africaines, pp. 395-426. In: C. Lévêque, M.N. Bruton and G.W.
40 956 Ssentongo (Eds.), Biologie et écologie des poissons d’eau douce africains. Coll. Travaux Doc.
957 216, ORSTOM, Paris.
958
959 Moreau, J., Christensen, V. and Pauly, D. 1993. A trophic ecosystem model of Lake
960 George, Uganda, pp. 124-129. In V.Christensen and D. Pauly (Eds.), Trophic models of
961 aquatic ecosystems. ICLARM Conference Proceedings 26, ICLARM, Manila, Philippines.
962
963 Moreau, J., Mavuti, K. and Daufresne, T. 2001. A synoptic Ecopath model of biomass flows
964 during two different static ecological situations in Lake Nakuru (Kenya). Hydrobiologia, 458:
965 63-74.
966
967 Moreau, J., Villanueva, M.C., Amarasinghe, U.S. and Schiemer, F. 2001. Trophic relationships
968 and possible evolution under various fisheries management strategies in Sri Lankan reservoir, pp.
969 201-214. In. S.S. De Silva and P. Lynch (Eds.), Fisheries management and Aquaculture in
970 Southeast Asian Lakes and Reservoirs. ACIAR Publications 98, Canberra, Australia.
971
972 Munyandero, J. and Mwape, L. 2003. Some records of Lake Tanganyika Clupeid and Latid fish
973 species, pp. 132-155. In M.L.D. Palomares, B. Samb, T. Diouf, J.M. Vakily and D. Pauly
974 (Eds.), Fish biodiversity: local studies as basis for global interferences. ACP-EU Fisheries
975 Research Report 14, Brussels.
976
977 Murichi, S.M., Hart, P.J.B. and Harper, D. 1995. The persistence of the two introduced species in
978 Lake Navaisha, Kenya, in face of environmental variability and fishing pressure, pp. 229-319. In
41 979 T. Pitcher ad P.J.B. Hart (Eds.), The impact of species changes in African Lakes. Fish and
980 Fisheries Series 18, Chapman and Hall, London.
981
982 Naeem, S., Knops, J.M.H., Tilman, D., Howe, K.M., Kennedy, T. and Gale, S. 2000. Plant
983 diversity increases resistance to invasion in the absence of covarying extrinsic factors. Oïkos, 91:
984 97-108.
985
986 Njiru, M., Okeyo-Owuor, J.B., Muchiri, M. and Cowx, I.G. 2004. Shifts in the food of the Nile
987 tilapia, Oreochromis niloticus (L.) in Lake Victoria, Kenya African Journal of Ecology, 42: 163-
988 170.
989
990 Nyström, P., Svensson, O., Lardner, B., Broenmark, C. and Graneli, W. 2001. The influence of
991 multiple introduced predators on a littoral pond community. Ecology, 82 (4): 1023-1039.
992
993 Odum, E.P. 1969. The strategy of ecosystem development. Science, 104: 262-270.
994
995 Odum, E.P. 1971. Fundamentals of ecology. W.B. Saunders Co, Philadelphia, 574 pp.
996
997 Odum, W.E. and Heald, E.J. 1975. The detritus-based food web of an estuarine mangrove
998 community, pp. 265-286. In: L.E. Cronin (Ed.), Estuarine research Vol. 1. Academic Press,
999 New York.
1000
1001 Ogutu-Ohwayo, R. 1990. The decline of the native fishes of Lake Victoria and Kyoga (East
1002 Africa) and the impact of introduced species, especially Nile Perch, Lates niloticus, and Nile
42 1003 tilapia, Oreochromis niloticus. Environmental Biology of Fishes, 27: 81-96.
1004
1005 Ogutu-Ohwayo, R., Hecky, R.E., Cohen, A.S. and Kaufman, L. 1997. Human impacts on the
1006 African Great Lakes. Environmental Biology of Fishes, 50: 117–131.
1007
1008 Olden, J.D. and Rooney, T.P. 2006. On defining and quantifying biotic homogenization. Global
1009 Ecology and Biogeography, 15: 113-120.
1010
1011 Pace, M.L., Cole, J.J., Carpenter, S.R. and Kitchell, J.F. 1999. Trophic cascades revealed in
1012 diverse ecosystems. Trends in Ecology and Evolution, 14: 483–488.
1013
1014 Palomares, M.L. and Pauly, D. 1998. Predicting food consumption of fish populations as
1015 functions of mortality, food type, morphometrics, temperature and salinity. Marine and
1016 Freshwater Research, 49: 447-453.
1017
1018 Pauly, D. 1980. On the interrelationships between natural mortality, growth parameters and
1019 mean environmental temperature in 175 fish stocks. Journal of the International Council for
1020 the Exploitation of the Sea, 39(3): 175-192.
1021
1022 Pauly, D. 1986. A simple method for estimating the food consumption of fish population from
1023 growth data and food conversion experiments. Fisheries Bulletin, 84: 827-840.
1024
1025 Pauly, D. and Palomares, M.L. 1987. Shrimp consumption by fishes 992 in Kuwait waters, a
1026 methodology, preliminary results and their implication for management and research. Kuwait
43 1027 Bulletin of Marine Sciences, 8: 101-125.
1028
1029 Pauly, D. and Christensen, V. 1995. Primary production required to sustain global fisheries.
1030 Nature, 374: 255-257.
1031
1032 Pauly, D., Christensen, V., Dalsgaard, J., Fröese, R. and Torres, F.C. Jr. 1998. Fishing down
1033 marine food webs. Science, 279: 860-863.
1034
1035 Pauly, D, Palomares, M.D. and Vakily, M. 2003. Trophic models of Northwest African marine
1036 ecosystems. In D. Pauly, M.D. Palomares and M. Vakily (Eds.), Marine Fisheries,
1037 Ecosystems and Societies in West Africa: Half a century of change, SIAP/EP/DT/03 24-28
1038 June 2002 Dakar, Senegal.
1039
1040 Pauly, D., Soriano-Bartz, M. and Palomares, M.L. 1993. Improved construction,
1041 parameterization and interpretation of steady-state ecosystem models, pp. 1-13. In V.
1042 Christensen and D. Pauly (Eds.), Trophic models of aquatic ecosystems. ICLARM Conference
1043 Proceedings 26, ICLARM, Manila.
1044
1045 Payne, I.A. 1986. The ecology of tropical lakes and rivers. Wiley & Sons, New York, 301 p.
1046
1047 Pearce, M.J. 1995. Effects of exploitation on the pelagic fish community in the south of Lake
1048 Tanganyika, pp. 425-441. In: T.J. Pitcher and P.J.B. Hart (Eds.), The impact of species
1049 changes in African Lakes. Fish and Fisheries Series 18 Chapman and Hall, London.
1050
44 1051 Perin, S., Pick, F.R., Lean, D.R.S. and Mazumder, A. 1996. Effects of planktivorous fish and
1052 nutrient additions in primary production of shallow versus deep (stratified) lake enclosures.
1053 Canadian Journal of Fisheries and Aquatic Sciences, 53: 1125-1132.
1054
1055 Persson, L. 1999. Trophic cascades: abiding heterogeneity and the trophic level concept at the
1056 end of the road. Oïkos, 85: 385–397.
1057
1058 Petchey, O.L., McPhearson, P.T., Casey, T.M. and Morin P.J. 1999. Environmental warming
1059 alters food-web structure and ecosystem function. Nature, 402: 69-72.
1060
1061 Pianka, E.R. 1973. The structure of lizard communities. Annual Review of Ecology and
1062 Systematics, 4: 53-74.
1063
1064 Pimentel, D., McNair, S., Janecka, J., Wightman, J., Simmonds, C., O’Connell, C., Wong, E.,
1065 Russel, L., Zern, J., Aquino, T. and Tsomondo, T. 2001. Economic and environmental threats of
1066 alien plant, animal, and microbe invasions. Agriculture, Ecosystems, and Environment, 84:1-
20. 1067 20.
1068
1069 Pitcher, T.J. and Hart, P.J.B. (Eds.). 1995. The impact of species changes in African Lakes.
1070 Kluwer Academic, Dordrecht, The Netherlands, 640 p.
1071
1072 Platts, T. and Ulanowicz, R.E. (Eds.) 1985. Ecosystem theory for biological oceanography.
1073 Canadian Bulletin of Fisheries and AquaticSciences, 213: 260 p.
1074
45 1075 Preikshot, D., Nsiku, E., Pitcher, T. and Pauly, D. 1998. An interdisciplinary evaluation of the
1076 status and health of African lake fisheries using a rapid appraisal. Journal of Fish Biology, 53
1077 (supplement A): 381-393.
1078
1079 Proulx, M., Pick, F.R., Mazumder, A., Hamilton, P.B. and Lean, D.R.S. 1996. Effects of
1080 nutrients and planktivorous fish on phytoplankton community structure versus deep (stratified)
1081 aquatic systems. Ecology, 77: 1556-1572.
1082
1083 Puth, L.M. and Post, D.M. 2005. Studying invasion: Have we missed the boat? Ecology Letters,
1084 8: 715-21
1085
1086 Quince, C., Higgs, P.G. and McKane, A.J. 2002. Food web structure and the evolution of
1087 ecological communities, pp. 281–298. In M. Lässig and A. Valleriani (Eds.), Biology
1088 Evolution and Statistical Physics. Springer, Berlin.
1089
1090 Raffaelli, D., van der Putten, W.H., Persson, L., Wardle, D.A., Petchey, O.L., Koricheva, J., van
1091 der Heijden, M., Mikola, J. and Kennedy, T. 2002. Multi-trophic dynamics and ecosystem
1092 processes, pp. 147-154. In M. Loreau, S. Naeem and P. Inchausti (Eds.), Biodiversity and
1093 ecosystem functioning: Synthesis and perspectives. Oxford University Press, Oxford.
1094
1095 Rejmánek, M.D. and Richardson, M. 1996. What attributes make some plant species more
1096 invasive? Ecology, 77: 1655-1661.
1097
1098 Richards, S.A., Nisbet, R.M., Wilson, W.G. and Possingham, H.P. 2000. Grazers and diggers:
46 1099 exploitation competition and coexistence among foragers with different feeding strategies on a
1100 single resource. The American Naturalist, 155: 266-279.
1101
1102 Rosenzweig, M.L. 2001. The four questions: what does the introduction of exotic species do to
1103 diversity? Evolutionary Ecology Research 3: 361-367.
1104
1105 Ruesink, J.L., Parker, I.M., Groom, M.J. and Kareiva, P.M. 1995. Reducing the risks of
1106 non indigenous species introductions. BioScience, 45 (7): 465-467.
1107
1108 Saltveit, S.J. 2006. The effects of stocking Atlantic salmon, Salmo salar, in a Norwegian
1109 regulated river. Fisheries Management and Ecology, 13: 197-205.
1110
1111 Sarmento, H., Isumbisho, P. and Descy, J.P. 2006. Phytoplankton ecology of Lake Kivu (Eastern
1112 Africa). Journal of Plankton Research, 28: 815-8829.
1113
1114 Sarvala, J., Salonen, K., Järvinen, M., Aro, E., Huttula, T., Kotilainen, P., Kurki, H., Langenberg,
1115 V., Mannini, P., Peltonen, A., Plisnier, P.D., Vourinen, I., Möslä, H. and Lindqvist, O. 1999.
1116 Trophic structure of Lake Tanganyika: carbon flows in the pelagic food web. Hydrobiologia,
1117 407: 149-173.
1118
1119 Sax, D.F. and Brown, J.H. 2000. The paradox of invasion. Global Ecology and Biogeography,
1120 9: 363-371.
1121
1122 Sax, D.F. and Gaines, S.D. 2003. Species diversity: from global decreases to local increases.
47 1123 Trends in Ecology and Evolution, 18 (11): 561-566.
1124
1125 Schutzenhofer, M.R. and Valone, T.J. 2006 Positive and negative effects of exotic Erodium
1126 cicutarium on an arid ecosystem. Biological Conservation, 132: 376-381.
1127
1128 Simberloff, D., Von Holle, B. 1999. Positive interactions of non-indigenous species: invasional
1129 meltdown? Biological Invasions, 1: 21-32.
1130
1131 Smith, S.D., Huwman, T.E., Zitzer, S.F., Charlet, T.N., Housman, D.C., Coleman, J.S.,
1132 Fenstermakerk, L.K., Seeman, J.R. and Nowak, R.S. 2000. Elevated CO2 increases productivity
1133 and invasive species success in an arid ecosystem. Nature, 408: 79-82.
1134
1135 Snoeks, J. 1994. The Haplochromines (Teleostei, Cichlidae) of Lake Kivu (East Africa): A
1136 taxonomic revision with notes on their ecology. Annales de Sciences Zoologiques, 270: 221-
1137 1102.
1138
1139 Snoeks, J., De Vos, L., Coenen, E. and Thys van den Audenaerde, D.T. 1990. Description de
1140 deux nouvelles espèces d’Haplochromis (Teleostei, Cichlidae) du Lac Kivu, Rwanda. Cybium,
1141 14(1): 63-76.
1142
1143 Spliethoff, P.C., De Iongh, H.H. and Frank, V.G. 1983. Success of the introduction of the
1144 freshwater clupeid Limnothrissa miodon (Boulenger) in Lake Kivu. Fisheries Management,
1145 14: 17-31.
1146
48 1147 Stachowicz, J.J., Fried, H., Whitlatch, R.B. and Osman, R.W. 2002. Biodiversity, invasion
1148 resistance and marine ecosystem function: reconciling pattern and process. Ecology, 83: 2575-
1149 2590.
1150
1151 Steiner, C.F. 2001. The effects of prey heterogeneity and consumer identity on the limitation
1152 of trophic-level biomass. Ecology, 82: 2495-2506.
1153
1154 Tadesse, Z. 1999. The nutritional status and digestibility of Oreochromis niloticus L. diet in Lake
1155 Langeno Ethiopia. Hydrobiologia, 416: 97-106.
1156
1157 Trewavas, E. 1983. Tilapiine fishes of the genera Sarotherodon, Oreochromis and Danakila.
1158 British Museum (Natural History), 654 p.
1159
1160 Twongo, T. 1995. Impact of fish species introductions on the tilapias of Lake Victoria and
1161 Kyoga, pp. 45-57. In: T.J. Pitcher and P.J.B. Hart (Eds.), The impact of species changes in
1162 African Lakes. Fish and Fisheries Series, 18 Chapman and Hall, London.
1163
1164 Ulanowicz, R.E. 1995. The part-whole relation in ecosystems, pp.549-560. In: B.C.
1165 Patten and S.E. Jørgensen (Eds.), Complex ecology. Prentice-Hall, Englewood Cliffs, New
1166 Jersey.
1167
1168 Ulanowicz, R.E. and Puccia, C.J. 1990. The mixed trophic impact routine. Coenose, 5: 7-16
1169
1170 Ulyel, A.P. 1991. Écologie alimentaire des Haplochromis spp. (Telepstei: Cichlidae) du Lac
49 1171 Kivu en Afrique Centrale. Thèse de Doctorat, Katholieke Universiteit Leuven, Département
1172 de Biologie, Écologie et de l’Aquaculture, 271 p.
1173
1174 Ulyel, A.P., Ollevier, F., Ceusters, R. and Thys van den Audenaerde, D.T. 1991. Food and
1175 feeding habits of Haplochromis (Teleostei: Cichlidae) from Lake Kivu (Central Africa) II. Daily
1176 feeding periodicity and dietary changes of some Haplochromis species under natural conditions.
1177 Belgian Journal of Zoology, 121 (2): 93-112.
1178
1179 Van den Bossche, J.P. and Bernacsek, G.M. 1990. Source book for the inland fishery resources
1180 of Africa: Volume 1, CIFA Technical Paper No.18.1, FAO, Rome, 240 pp.
1181
1182 Van der Zanden, M.J., Casselman, J.M. and Rasmussen, J.B. 1999. Stable isotope evidence for
1183 the food web consequences of species invasions in lakes. Nature, 401: 464–467.
1184
1185 Verheyen, E., Salzburger, W., Snoeks, J. and Meyer, A. 2003. Origin of the superflock of cichlid
1186 fishes from Lake Victoria, East Africa. Science, 300: 325-328.
1187
1188 Villanueva, M.C., Lalèyè, P., Albaret, J.-J., Laë, R., Tito de Morais, L. and Moreau, J. 2006.
1189 Comparative analysis of trophic structure and interactions of two tropical lagoons. Ecological
1190 Modelling, 197: 461-477.
1191
1192 Villanueva, M.C. and Moreau, J. 2001. Recent trends of evolution in Lake Victoria fisheries as
1193 assessed by Ecopath 4 (with Ecosim) software, pp. 96-110. In: I.G. Cowx (Ed.), Lake and
1194 Reservoirs Fisheries Management. The University of Hull Publication, United Kingdom.
50 1195
1196 Vitousek, P.M., D'Antonio, C.M., Loope, L.L., Rejmánek, M. and Westbrooks, R. 1997.
1197 Introduced species: A significant component of human-caused global change. New Zealand
1198 Journal of Ecology, 21(1): 1-16.
1199
1200 Volterra, V. 1931. Variations and fluctuations of the number of individuals in animal species
1201 living together, 409-448 pp. In: R.N. Chapman (Ed.), Animal Ecology. McGraw-Hill, New York
1202 and London.
1203
1204 Wanink, J.H. and Witte, F. 2000. The use of perturbation as a natural experiment: effects of
1205 predator introduction on the community structure of zooplanktivorous fish in Lake Victoria.
1206 Advances in Ecological Research, 31: 553-570.
1207
1208 Wanink, J.H., Katunzi, E.F.B., Goudswaard, K.P.C., Witte, F. and van Densen, W.L.T. 2002.
1209 The shift to smaller zooplankton in Lake Victoria cannot be attributed to the ‘sardine’
1210 Rastrineobola argenta (Cyprinidae). Aquatic Living Resources, 15: 37-43.
1211
1212 Welcomme, R.L. 1988. International transfers of inland fish species. FAO Fisheries Technical
1213 Papers 294, 318 p.
1214
1215 Wilcove, D.S., Rothstein, D., Dubow, J., Phillips, A. and Losos, E. 1998. Quantifying threats to
1216 imperilled species in the United States. Bioscience, 48: 607-615.
1217
51 1218 Winfield, I.J. and Durie, N.C. 2004. Fish introductions and their management in the English
1219 Lake District. Fisheries Management and Ecology, 11: 195-201.
1220
1221 Wootton, J.T., Power, M.E., Paine, R.T. and Pfister, C.A. 1996. Effects of productivity,
1222 consumers, competitors, and El Niño events on food chain patterns in a rocky intertidal
1223 community. Proceedings of the National Academy of Sciences, 93: 13855–13858.
1224
1225 Yang, Y.F., Huang, X.F., Liu, J.K. and Jiao, N.Z. 2005. Effects of fish stocking on the
1226 zooplankton community structure in a shallow lake in China. Fisheries Management and
1227 Ecology, 12: 81-89.
1228
1229 Yodzis, P. and Winemiller, K.O. 1999. In search of operational trophospecies in a tropical
1230 aquatic food web. Oïkos, 87: 327-340.
52 Figure
List of Figures:
Figure 1. Lake Kivu (East Africa) on the border between Republic Democratic of Congo (formerly known as Zaire) and Rwanda. The sector considered in the study is the southern basin of the lake (inside dashed frame).
Figure 2. Relative biomass and major flows connecting of functional groups considered in the Lake Kivu model. Less important flows are omitted for clarity’s sake. The horizontal axis of symmetry of each box is aligned with the trophic level (TL) of the box in question. The numbered value of a TL is fractional because it depends on the diet composition of this group and on the TLs of its preys (Christensen and Pauly, 1993).
Figure 3. Prey versus predator niche overlap plot. Groups in the lower left of the figure have quite no overlap and are quite independent for both preys and predators. Groups on the upper right corner have a high overlap for both predators and preys
Figure 4. Combined trophic impacts for functional groups considered. Clear box: cumulative effects (absolute values) of an increase by 10% in biomass of all other groups to a specific group. Shaded box: sum of the absolute values of impacts of this group on other groups. Positive impacts are shown above the base line, negative impacts are shown below.
Figure 5. Cumulative impacts of fishing gear on functional groups if effort is increased by 10%. Positive impacts are shown above the base line, negative impacts are shown below.
Goma
N
Gisenyi
R E P U B L I C NORTHERN D E M O C R A T I C BASIN O F C O N G O
R W A N D A
KALEHE BASIN EASTERN BASIN Kibuye
IDJWI ISLAND
Bukavu 0 20 km BUKAVU BASIN Cyangugu FISHERIES CENTER
Figure 1
Villanueva et al. 4 R. moorei
H. vittatus Clarias spp. 0.01 0.01 0.03 0.1 B = 0.03 Haplochromis 0.2 P = 0.04 B = 0.03 0.1 Adult B = 0.20 benthophages P = 0.04 Q = 0.3 L. miodon P = 0.10 0.77 Q = 0.3 Barbus spp. 1.16 Q = 0.9 B = 0.53 0.01 2.7 Juvenile L. miodon 0.1 P = 1.89 5.7 B = 1.32 3.3 1.66 3 P = 2.32 B = 0.02 Q = 0.3 B = 0.50 P = 0.04 Q = 22.8 P = 3.32 Q = 0.4
Q = 15.1 Benthic fauna 61.3 Other Tilapias 0.15 0.07 Zooplankton B = 3.68 O. niloticus 2.4 B = 0.33 658.3 P = 10.45 B = 0.38 3.2 TrophicTrophic Level Level P = 1.18 0.07 P = 0.46 B = 0.50 Q = 165.4 B = 0.37 Q = 11.8 P = 3.32 2 2.1 Q = 10.2 97.2 P = 0.37 Haplochromis Q = 1944.0 Q = 9.3 piscivores
1002.6
Phytoplankton Detritus Legend 1 B = 34.17 P = 2733.84 B = 165.00 Catch Flows back to detritus Figure 2 Cannibalism
Villanueva et al.
6, 7 9, 10 Legend 1 A C 5, 9 5, 10 1 Clarias spp 2, 3 2, 8 2 R. moorei 2, 4 3 H. vittatus 3, 8 7, 11 4 Haplochromis benthophages 5, 1 5 Haplochromis planktivores 3, 4 11, 12 6 Adult L. miodon 5, 7 3, 6 7 Juvenile L. miodon 6, 8 B D 4, 6 8 Barbus spp 7, 10 7, 12 9 O. niloticus PreyPreyPrey overlap overlap overlap 3, 7 7, 9 5, 6 2, 6 8, 10 10 Other Tilapiines 7, 8 11 Benthic fauna 6, 10 4, 7 4, 10 3, 10 12 Zooplankton 5, 8 8, 9 4, 11 2, 10 3, 5 6, 9 2, 7 3, 9 4, 5
0 1 Predator overlap
Figure 3
Villanueva et al.
3
2
1 0
-1
-2 -3 Clarias spp. R. moorei H. vittatus Haplochromis Haplochromis Adult L. Juvenile L. Barbus spp. O. niloticus Other Benthic fauna Zooplankton Phytoplankton Detritus benthophages planktivores miodon miodon Tilapiines
Impacting group Impacted group
Figure 4
Villanueva et al. 0.6 Trimaran Gillnet Beach Seine Longline 0.5
0.4
0.3
0.2
0.1
0
-0.1 Clarias spp. R. moorei H. vittatus Haplochromis Haplochromis Adult L. Juvenile L. Barbus spp. O. niloticus Other Benthic fauna Zooplankton Phytoplankton Detritus benthophages planktivores miodon miodon Tilapiines -0.2
Figure 5
Villanueva et al. Tables
Table 1. Growth parameters for fish populations as incorporated in the model.
Species L∞ K Z M (TL cm) (yr-1) (yr-1) (yr-1) Clarias gariepinus 102.8 a 0.165 a 0.500 a 0.363 c Raiamas moorii 23.0 b 0.680 b 1.640 b 1.389 c Haplochromis vittatus 32.5 b 0.550 b 1.450 b 1.100 c Haplochromis graueri 17.2 b 1.085 b 3.550 b 2.049 c Haplochromis nigroides 10.7 b 1.682 b 3.580 b 3.115 c Adult Limnothrissa miodon 18.0 b 1.100 b 1.750 b 2.147 c Juvenile Limnothrissa miodon 15.0 b 1.100 b 6.690 b 2.040 c Barbus kerstenii 11.7 a 1.330 a 2.786 b 2.601 c Oreochromis niloticus eduardianus 38.7 a 0.550 a 1.008 a 0.806 c Oreochromis macrochir 28.3 a 0.653 a 1.200 a 1.007 c a Estimated based on models of de Merona (1984) and Fröese and Binohlan (2000) based on observed maximum total length b Computed using the FiSAT software (Gayanilo et al., 2002); c Calculated using the predictive formula of Pauly (1980).
Table 2. Input values and calculated parameters (in bold) for the Ecopath model of the Congolese sector of Lake Kivu. TL is the trophic level, HA is the habitat area (%), BHA is the biomass calculated for the habitat area (t.km-2), B is the total biomass (t.km-2), P/B is the production rate (yr-1), Q/B the annual foof consumption per unit biomass (yr-1), Y the total catch (t.km-2 yr-1).and EE is the ecotrophic efficiency .
d Group name TL HA B HA B P/B Q/B Y EE Clarias spp 3.36 1.0 0.204 0.204a 0.50c 4.30e 0.028 0.275 Raiamas moorei 3.56 1.0 0.025 0.025 1.64c 21.03f 0.010 0.500d Haplochromis vittatus 3.42 1.0 0.029 0.029a 1.45c 10.04e 0.010 0.797 Haplochromis benthivores 3.15 0.1 5.320 0.532a 3.55c 17.56e 0.773 0.544 Haplochromis planktivores 2.20 1.0 0.329 0.329a 3.58c 35.80e 0.153 0.279 L. miodon adults 3.04 1.0 1.324 1.324b 1.75c 17.20f 1.159 0.511 L.miodon juveniles 2.80 1.0 0.496 0.496b 6.69c 30.36f 1.659 0.917 Barbus spp 2.98 0.1 0.156 0.016 2.79c 27.86e 0.010 0.500d O. niloticus eduardianus 2.00 0.1 3.670 0.367a 1.01c 25.39e 0.075 0.479 Other Tilapiine fish 2.18 0.1 3.830 0.383a 1.20c 26.74e 0.074 0.233 Benthic fauna 2.32 0.1 36.757 3.676 4.50d 45.00d 0.900d Zooplankton 2.05 1.0 10.800 10.800 26.00d 180.00d 0.626 Phytoplankton 1.0 34.173 34.173 80.00d - 0.633 Detritus 1.0 165.00 165.00d - - 0.151
a Biomass (B) was estimated when possible directly from the ratio between catch (Y) and annual fishing mortality as computed in Ecopath (B = Y/F); b Based on echo sound data by Lambœuf (1991). Biomass of adults are higher (1.324 t.km-2) than those of their juveniles (0.496 t.km-2) due to the demographical structure of the population. c P/B of functional group is assumed equal to the total mortality estimated (cf. Table 1); d Sources indicated in text; e Estimated using the predictive model of Palomares and Pauly (1998); f Estimated using the MAXIMS software (Jarre et al., 1991).
Villanueva et al. MS 5469A revision
Table 3. Diet composition (%) of the groups considered in the Ecopath model of the Lake Kivu Congolese sector.
Group name 1 a 2 b 3 c 4 d 5 e 6 b 7 f 8 g 9 h 10 i 11 j 12 k 1 Clarias spp 2 Raiamas moorei 0.005 0.020 3 Haplochromis vittatus 0.015 0.02 4 Haplochromis benthivores 0.120 0.15 0.200 0.03 5 Haplochromis planktivores 0.090 0.10 0.080 0.05 6 L. miodon adult 0.020 0.025 7 L.miodon juvenile 0.100 0.25 0.100 0.05 8 Barbus spp 0.010 0.010 9 O. niloticus eduardianus 0.100 0.050 10 Other Tilapiine fish 0.030 0.025 11 Benthic fauna 0.400 0.46 0.270 0.78 0.03 0.20 0.05 0.57 0.10 12 Zooplankton 0.040 0.01 0.150 0.12 0.15 0.65 0.70 0.10 0.05 0.30 0.05 13 Phytoplankton 0.020 0.01 0.020 0.01 0.8 0.09 0.23 0.05 0.80 0.65 0.30 0.85 14 Detritus 0.050 0.050 0.09 0.02 0.01 0.02 0.20 0.20 0.20 0.40 0.10
a. Diet composition based on indications of Mavuti et al. (1996) and from Fröese and Pauly (2006); b. From Kaningini et al. (1999); c. Ulyel (1991) and Kaningini et al. (1999); d. Considered diet composition of Haplochromis graueri (Ulyel, 1991); e. Considered diet composition of Haplochromis kamiranzovu (Ulyel, 1991); f. Kaningini et al. (1999) and Roest (1999); g. M. Isumbisho (UERAH, unpublished data); ;(..h. From Trawavas (1983) and B Kaningini (UERAH, pers. ٛ omm i. Considered diet composition of O. macrochir from Fröese and Pauly (2006); j. Based on indications of Christensen and Pauly (1993); Table 52. From Christensen and Pauly (1993) and Sarvala et al. (2003).
Table 4. Ecological characteristics of the groups considered in the model. P/Q is the production/consumption ratio, FtD is the flow to detritus (t km-2yr-1), NE is the net efficiency and OI is the omnivory index.
Group name P/Q FtD NE OI Clarias spp 0.116 0.249 0.145 0.282 Raiamas moorei 0.078 0.124 0.097 0.150 Haplochromis vittatus 0.144 0.067 0.181 0.339 Haplochromis benthivores 0.202 2.730 0.253 0.155 Haplochromis planktivores 0.100 3.205 0.125 0.179 L. miodon adult 0.102 5.688 0.127 0.153 L.miodon juvenile 0.220 3.286 0.275 0.218 Barbus spp 0.100 0.108 0.125 0.349 O. niloticus eduardianus 0.040 2.057 0.050 0.000 Autres Tilapiine fish 0.045 2.401 0.056 0.195 Benthic fauna 0.100 51.277 0.143 0.233 Zooplankton 0.144 688.307 0.206 0.053 Phytoplankton - 1002.627 - 0.000 Detritus - 0.000 - 0.298
Villanueva et al. MS 5469A revision Table 5. The trophic structure of the Lake Kivu ecosystem (Congolese sector) as estimated by the Ecopath software.
Trophic level Catch (%) per Biomass % per TL Transfer efficiency (TL) (t km-2 yr-1) TL (t km-2) (%) V 0.0031 0.08 0.012 0.32 4.60 IV 0.3405 8.62 0.365 9.86 7.60 III 2.7310 69.12 2.034 54.93 9.40 II 0.8767 22.19 1.292 34.89 4.50 Proportion of total flow originating from detritus: 0.31 Transfer efficiencies: 1) From primary producers: 6.80 % 2) From detritus: 7.10 % Total: 6.90 %
Table 6. Summary statistics of Lake Kivu.
Parameter Value Units Sum of all consumption 2190.04 t.km-².yr-1 Sum of all exports 1499.85 t.km-².yr-1 Sum of all respiratory flows 1233.99 t.km-².yr-1 Sum of all flows into detritus 1762.12 t.km-².yr-1 Total system throughput 6686.00 t.km-².yr-1 Sum of all production 3040.00 t.km-².yr-1 Mean trophic level of the catch 2.90 Gross efficiency (catch/net p.p.) 0.001445 Input total net primary production 2733.84 t.km-².yr-1 Calculated total net primary production 2733.84 t.km-².yr-1 Total primary production/total respiration 2.21 Net system production 1499.85 t.km-².yr-1 Total primary production/total biomass 52.22 Total biomass/total throughput 0.008 Total biomass (excluding detritus) 52.35 t.km-² Total catch 3.95 t.km-².yr-1 Connectance Index (CI) 0.396 System Omnivory Index (SOI) 0.148
Villanueva et al. MS 5469A revision